AFFILIATIONS
1 Emergency Department, Henri-Mondor University Hospital, Greater Paris University Hospitals, Créteil, France, Créteil, France, Laboratoire d’Excellence, INSERM U955, IMRB, Team Bartolucci, Créteil, France, Créteil, France
2 Laboratoire d’Excellence, INSERM U955, IMRB, Team Bartolucci, Créteil, France, Créteil, France, Service de médecine interne, Centre national de référence des cytopénies auto-immunes de l'adulte, Hôpital Henri Mondor, AP–HP, Créteil, France, Créteil, France
3 Department of Biochemistry and Molecular Biology, Henri Mondor University Hospital, Créteil, France, Créteil, France
4 Department of Biochemistry and Molecular Biology, Henri Mondor University Hospital, Créteil, France, Créteil, France
5 Service d'Accueil des Urgences, Centre Hospitalier Sud Francilien, corbeil-Essonnes, France
6 Department of Biochemistry and Molecular Biology, Henri Mondor University Hospital, Créteil, France, Créteil, France
7 Laboratoire d’Excellence, INSERM U955, IMRB, Team Bartolucci, Créteil, France, Créteil, France
8 Laboratoire d’Excellence, INSERM U955, IMRB, Team Bartolucci, Créteil, France, Créteil, France
9 Centre de référence des syndromes drépanocytaires majeurs, Hôpital Henri-Mondor, AP–HP, Créteil, France, Créteil, France
10 Platform of Biological Resources, Henri Mondor Hospital, Créteil, France, Créteil, France
11 Sickle Cell Disease Referral Center, Avicenne Hospital, Assistance Publique Hôpitaux de Paris (AP-HP), Bobigny, France. Clinical Research Unit, Mayotte Hospital, Mamoudzou, France. IAME, INSERM, UMR 1137, University Sorbonne Paris-Nord, IMRB Mondor, Créteil., Créteil, France
12 Laboratoire d’Excellence, INSERM U955, IMRB, Team Bartolucci, Créteil, France, Créteil, France, Service de médecine interne, CHU Avicenne, AP-HP, Bobigny, France, Créteil, France
13 Centre de référence des syndromes drépanocytaires majeurs, Hôpital Henri-Mondor, AP–HP, Créteil, France, Créteil, France, Laboratoire d’Excellence, INSERM U955, IMRB, Team Bartolucci, Créteil, France, Créteil, France, University Paris-Est Créteil, Créteil, France, Créteil, France
14 Établissement Français du Sang, Île-de-France Mondor, Créteil, France, Créteil, France, Laboratoire d’Excellence, INSERM U955, IMRB, Team Bartolucci, Créteil, France, Créteil, France
ABSTRACT
Background
Acute chest syndrome (ACS) is a major cause of morbidity and mortality in sickle cell disease (SCD) and complicates approximately 15–20% of vaso-occlusive crises (VOC), often within 2–3 days after admission (Kassasseya et al., 2026; Cheminet et al., 2023; Bartolucci et al., 2016). Although the PRESEV score reliably identifies patients at low risk, its positive predictive performance remains limited (Bartolucci et al., 2016; Kassasseya et al., 2026). Inflammation is central to VOC pathophysiology (Kato et al., 2017), but the longitudinal inflammatory profile preceding ACS in adults remains insufficiently characterized. We aimed to describe the trajectories of circulating inflammatory mediators during VOC and determine whether cytokine profiling could improve the early identification of patients at risk of developing ACS.
Methods
We included patients from the single-center PRESEV1 cohort conducted at Henri-Mondor University Hospital between 2006 and 2012 who had serum available at admission Day 1 (D1), D2, D4, and steady state (≥30 days after the acute event and ≥3 months after a transfusion). All patients had hemoglobin SS or S-β0 thalassemia and were hospitalized for severe VOC requiring opioid analgesia, without ACS at emergency department presentation.
Routine inflammatory biomarkers were measured in the hospital biochemistry laboratory: IL-6 by electrochemiluminescence immunoassay; calprotectin by a particle-enhanced immunoturbidimetric immunoassay; CRP, orosomucoid, and haptoglobin by immunoturbidimetry; and serum amyloid A by immunoturbidimetry on an IMMAGE® system. A total of 33 cytokines involved in SCD-related vaso-occlusion and ACS were analyzed using two Thermo Scientific multiplex panels.
Concentrations were transformed using ln(x+1) to reduce right skewness and the influence of extreme values. Groups were compared using Mann–Whitney tests, and longitudinal trajectories were assessed using linear mixed-effects models. D1 discrimination was evaluated by ROC analysis, followed by exploratory multivariable logistic regression. Multiple testing was controlled using the Benjamini–Hochberg false discovery rate (FDR).
Results
Of the 247 patients enrolled, 55 had serum available at all four time points: 40 had uncomplicated VOC and 15 subsequently developed ACS, which occurred at a median of day 3 [IQR, 2–4]. Patients who developed ACS tended to be younger than those with uncomplicated VOC (23 years [IQR, 20–29] vs. 26 years [IQR, 23–31.5]; p=0.09), while sex distribution was comparable (53.8% vs. 61.5% male; p=0.7). Most mediators were comparable at steady state, although IL-2, IL-6, and IL-12p70 were lower in patients who subsequently developed ACS.
In patients with uncomplicated VOC, concentrations increased relative to steady state from D1 for CRP, IL-6, serum amyloid A, calprotectin, and orosomucoid; at D2 for GM-CSF and haptoglobin. In contrast, LIF decreased from D2, while IL-12p70, IL-17A, MIP-1α, TNF-α, and IL-6RA decreased at D4. ACS exhibited distinct inflammatory trajectories compared to uncomplicated VOC at D1 in IL-2, IL-6, IL-7, and IL-13; at D2 in CRP, IL-12p70, LIF, and serum amyloid A; and at D4 in orosomucoid, IL-17A, and IL-5. Several mediators showed opposite changes between the groups, decreasing in uncomplicated VOC but increasing in patients who developed ACS. After FDR adjustment, only the D2 orosomucoid difference remained significant; all other results were exploratory.
We then investigated whether D1 biomarkers could identify patients at risk of subsequently developing ACS. The highest univariate AUCs were observed for MCP-1 (0.69), IL-6 (0.67), and IL-7 (0.65), although none reached statistical significance. An exploratory model combining CRP, MCP-1, and IL-7 achieved an apparent AUC of 0.929, with apparent PPV and NPV of 100% and 92%, respectively.
Conclusions and perspectives
Longitudinal inflammatory profiling identified distinct trajectories in patients who subsequently developed ACS, with the earliest between-group differences involving IL-2, IL-6, IL-7, and IL-13. An exploratory model combining CRP, MCP-1, and IL-7 improved early ACS discrimination. These findings require validation in larger independent cohorts. If confirmed, cytokine profiling could complement the widely accessible PRESEV score, which retains strong performance for ruling out ACS. The early IL-6 signal also supports further evaluation of IL-6-related pathways as potential therapeutic targets during VOC.